Method for machine learning using online convex optimization problem solving with minimum regret
    1.
    发明授权
    Method for machine learning using online convex optimization problem solving with minimum regret 失效
    使用在线凸优化问题求解的机器学习方法以最小的遗憾

    公开(公告)号:US07870082B2

    公开(公告)日:2011-01-11

    申请号:US11759505

    申请日:2007-06-07

    IPC分类号: G06F15/18

    摘要: Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.

    摘要翻译: 为在线凸优化问题提供了方法,系统和计算机程序产品,其中决策者了解所有过去状态,并为其以前的选择产生成本函数,并尝试进行最终遗憾的新选择。 该方法不依赖于成本函数的结构或状态的表征,并且利用连续状态之间的相似性来使得该方法能够收敛到合理的最佳结果。

    METHOD FOR MACHINE LEARNING WITH STATE INFORMATION
    2.
    发明申请
    METHOD FOR MACHINE LEARNING WITH STATE INFORMATION 失效
    用国家信息机器学习的方法

    公开(公告)号:US20080306887A1

    公开(公告)日:2008-12-11

    申请号:US11759505

    申请日:2007-06-07

    IPC分类号: G06F15/18

    摘要: Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.

    摘要翻译: 为在线凸优化问题提供了方法,系统和计算机程序产品,其中决策者了解所有过去状态,并为其以前的选择产生成本函数,并尝试进行最终遗憾的新选择。 该方法不依赖于成本函数的结构或状态的表征,并且利用连续状态之间的相似性来使得该方法能够收敛到合理的最佳结果。

    Fast adaptation in real-time systems
    3.
    发明授权
    Fast adaptation in real-time systems 失效
    快速适应实时系统

    公开(公告)号:US08494994B2

    公开(公告)日:2013-07-23

    申请号:US12826672

    申请日:2010-06-30

    IPC分类号: G06F17/00

    CPC分类号: G06F17/11

    摘要: According to one embodiment of the present invention, a method for online convex optimization is provided. The method includes performing a step that relies on the selection of x at a time t (xt), where x is a variable involved with the step. A resulting cost (ƒt(xt)) is calculated that results from selecting xt when performing the step, where ƒt is a cost function. A minimum possible cost (ƒt(x*t)) associated with the selection of x* is then found and the difference between the resulting cost (ƒt(xt)) and the minimum possible cost (ƒt(x*t)) is determined. A direction of movement from xt to xt+1 is selected and a subsequent step that relies on the section of xt+1 is then performed.

    摘要翻译: 根据本发明的一个实施例,提供一种在线凸优化的方法。 该方法包括执行在时间t(xt)处依赖于x的选择的步骤,其中x是该步骤涉及的变量。 计算得出的成本(ft(xt)),这是在执行步骤时选择xt的结果,其中ft是成本函数。 然后找到与选择x *相关联的最小可能成本(ft(x * t)),并确定所得成本(ft(xt))与最小可能成本(ft(x * t))之间的差异 。 选择从xt到xt + 1的移动方向,然后执行依赖于xt + 1的部分的后续步骤。

    FAST ADAPTATION IN REAL-TIME SYSTEMS
    4.
    发明申请
    FAST ADAPTATION IN REAL-TIME SYSTEMS 失效
    快速适应实时系统

    公开(公告)号:US20120005142A1

    公开(公告)日:2012-01-05

    申请号:US12826672

    申请日:2010-06-30

    IPC分类号: G06N5/02

    CPC分类号: G06F17/11

    摘要: According to one embodiment of the present invention, a method for online convex optimization is provided. The method includes performing a step that relies on the selection of x at a time t (xt), where x is a variable involved with the step. A resulting cost (ƒt(xt)) is calculated that results from selecting xt when performing the step, where ƒt is a cost function. A minimum possible cost (ƒt(x*t)) associated with the selection of x* is then found and the difference between the resulting cost (ƒt(xt)) and the minimum possible cost (ƒt(x*t)) is determined. A direction of movement from xt to xt+1 is selected and a subsequent step that relies on the section of xt+1 is then performed.

    摘要翻译: 根据本发明的一个实施例,提供一种在线凸优化的方法。 该方法包括执行在时间t(xt)处依赖于x的选择的步骤,其中x是该步骤涉及的变量。 计算得出的成本(ƒt(xt)),这是在执行步骤时选择xt的结果,其中ƒt是成本函数。 然后找到与选择x *相关联的最小可能成本(ƒt(x * t)),并确定所得成本(ƒt(xt))与最小可能成本(ƒt(x * t))之间的差异 。 选择从xt到xt + 1的移动方向,然后执行依赖于xt + 1的部分的后续步骤。

    METHOD FOR MACHINE LEARNING WITH STATE INFORMATION

    公开(公告)号:US20080306891A1

    公开(公告)日:2008-12-11

    申请号:US12134073

    申请日:2008-06-05

    IPC分类号: G06F15/18

    摘要: Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.

    Method and apparatus for collaborative selection of proposals
    7.
    发明授权
    Method and apparatus for collaborative selection of proposals 有权
    合作选择提案的方法和设备

    公开(公告)号:US08489443B2

    公开(公告)日:2013-07-16

    申请号:US12051783

    申请日:2008-03-19

    申请人: Nimrod Megiddo

    发明人: Nimrod Megiddo

    IPC分类号: G06Q10/00

    CPC分类号: G06Q10/10

    摘要: A method and apparatus for selecting a subset of proposals from a set of proposals submitted via a web site, including defining proposal attributes for the selecting, formulating selection requirements for the subset, establishing metrics for comparing proposals according to the proposal attributes and the selection requirements, evaluating each of the set of proposals according the established metrics, distributing the results of evaluating each of the set of proposals to the submitter of each proposal, formulating an optimization problem for determining the subset of proposals, determining the subset of proposals by solving the optimization problem, and accepting feedback about the determining including changing the optimization problem based on the feedback.

    摘要翻译: 一种用于从通过网站提交的一组提案中选择提案子集的方法和装置,包括定义用于选择的提案属性,为该子集制定选择要求,根据提案属性和选择要求建立用于比较提案的度量 根据既定的指标评估每组提案,将每一个提案的评估结果分配给每个提案的提交者,制定用于确定提案子集的优化问题,通过解决方案来确定提案的子集 优化问题,并接受关于确定的反馈,包括基于反馈来改变优化问题。

    Method and system for authenticating telephone callers and avoiding unwanted calls
    8.
    发明授权
    Method and system for authenticating telephone callers and avoiding unwanted calls 有权
    用于认证电话呼叫者并避免不必要呼叫的方法和系统

    公开(公告)号:US08467512B2

    公开(公告)日:2013-06-18

    申请号:US12512347

    申请日:2009-07-30

    IPC分类号: H04M1/56

    CPC分类号: H04M15/06

    摘要: A service that handles incoming telephone calls without bothering the telephone subscriber is disclosed. The service permits a call to go through to a subscriber if the service determines that the call is not unwanted and the caller has been unauthenticated. The authentication is based on challenging the caller to prove its identity rather than relying on caller ID displays. Prospective callers pre-register with the service providing caller account information. When a caller is issued a challenge, the caller may prove its authenticity by supplying the challenge back to the service along with its registered information.

    摘要翻译: 公开了一种处理传入电话而不打扰电话用户的服务。 如果服务确定呼叫不是不需要的,并且呼叫者未经身份验证,则该服务允许呼叫转到订户。 认证是基于挑战来电者来证明其身份,而不是依靠来电显示。 预期来电者预先注册提供来电者帐户信息的服务。 当呼叫者发出挑战时,呼叫者可以通过向服务提供质询及其注册信息来证明其真实性。

    Method for creating and maintaining threads of phone/email/fax/SMS conversations
    9.
    发明授权
    Method for creating and maintaining threads of phone/email/fax/SMS conversations 失效
    创建和维护电话/电子邮件/传真/短信对话线程的方法

    公开(公告)号:US07539295B1

    公开(公告)日:2009-05-26

    申请号:US12195904

    申请日:2008-08-21

    IPC分类号: H04M3/00

    摘要: A method for creating and maintaining threads of communications comprises accepting an electronic-communication call from a caller, determining if the caller is a new or previous caller by determining if a prior-caller user identification is contained within a call-receiving system, responsive to determining that the caller is a new caller assigning a new-caller user identification to the new caller and assigning a unique thread identification for communications related to a new issue, and responsive to determining that the caller is a known previous caller verifying a known-caller user identification and determining whether the known previous caller is calling about a new issue. Responsive to determining that the known previous caller is calling about a new issue, the method further includes assigning a unique thread identification to the new issue. Responsive to determining that the known previous caller is calling about an existing issue, the method further includes obtaining a list of previous thread identifications.

    摘要翻译: 用于创建和维护通信线程的方法包括接受来自呼叫者的电子通信呼叫,通过确定呼叫接收系统中是否包含先前呼叫者用户标识来确定呼叫者是新呼叫者还是先前呼叫者,响应于 确定呼叫者是新呼叫者,向新呼叫者分配新的呼叫者用户标识,并为与新问题相关的通信分配唯一的线程标识,并且响应于确定呼叫者是已知的以前的呼叫者来验证已知呼叫者 用户识别并确定已知的先前呼叫者是否在呼叫新的问题。 响应于确定已知的先前呼叫者正在调用新的问题,该方法还包括向新问题分配唯一的线程标识。 响应于确定已知的先前呼叫者正在调用现有问题,该方法还包括获得先前线程标识的列表。

    DISCOVERING INTERESTINGNESS IN FACETED SEARCH
    10.
    发明申请
    DISCOVERING INTERESTINGNESS IN FACETED SEARCH 审中-公开
    发现面对搜索的兴趣

    公开(公告)号:US20090106244A1

    公开(公告)日:2009-04-23

    申请号:US12200981

    申请日:2008-08-29

    IPC分类号: G06F17/30

    摘要: Exemplary embodiments of the present invention relate to enhanced faceted search support for OLAP queries over unstructured text as well as structured dimensions by the dynamic and automatic discovery of dimensions that are determined to be most “interesting” to a user based upon the data. Within the exemplary embodiments “interestingness” is defined as how surprising a summary along some dimensions is from a user's expectation. Further, multi-attribute facets are determined and a user is optionally permitted to specify the distribution of values that she expects, and/or the distance metric by which actual and expected distributions are to be compared.

    摘要翻译: 本发明的示例性实施例涉及通过基于数据被确定为对用户最“有趣”的维度的动态和自动发现的针对非结构化文本的OLAP查询的增强的分面搜索支持以及结构化维度。 在示例性实施例中,“兴趣”被定义为从用户的期望来看,某些维度上的摘要令人惊讶。 此外,确定多属性小平面,并且可选地允许用户指定她期望的值的分布,和/或要比较实际和预期分布的距离度量。